The maintenance of stable neuron morphologies and connectivity patterns is crucial for maintaining proper neural circuit function. This is particulary relevant for neurons in the visual system, which are highly dynamic and yet maintain stable connectivity patterns that allow visual information collected by photoreceptors in the retina to be reliably transmitted to subsequent visual ganglia by retinotopic mapping. However, the physiological mechanisms that control long-term morphological homeostasis in the visual system are almost completely mysterious. The highly-stereotyped visual system of the fruit fly, Drosophila melanogaster, provides an excellent system for dissecting the molecular and cellular mechanisms that control the activity-dependent maintenance and adaptation of neuron morphology. Moreover, powerful genetic tools, sophisticated behavioral assays, and well-established in vivo imaging techniques make Drosophila ideal not only for investigating the mechanisms that control neuron morphological homeostasis, but also for studying the manner in which morphological adaptations contribute to neural circuit activity and visually-motivated fly behaviors. This proposal will address these questions by correlating quantitative measurements of neuron morphology with measurements of sub- cellular molecular localization patterns, as well as measurements of large-scale neural activity patterns and fly behaviors. In addition, the morphological and behavioral consequences of photoreceptor activity perturbations, as well as the effects of direct perturbations of molecular activities and localization patterns, will be assessed. These experiments will provide a molecular framework for understanding the activity-dependent mechanisms that control morphological homeostasis and adaptation in the Drosophila visual system. Moreover, this proposal will begin to elucidate the relationship between neural form and function, linking cellular mechanisms to large-scale animal behavior.